All Projects

If you boot the system with the lid closed, the backlight will still be on when you get to the desktop. You have to lift the lid for an instant and then close again to shut it off. Perhaps this is because of the lack of trigger for an interrupt since the sensor is already activated. Can there be one extra check of the lid sensor after boot to cover this static position?

Battery remaining percentage in MiniMenu does not update unless you change tabs back and forth.

This is only really an issue when leaving the unit charging and not using it. It gives the appearance that it is not charging until you swap tabs around.

PyGame is damn slow (games with scrolling are not at all playable).
Would be nice if Hardware surface could be used :)

The title says everything: The 'del' key does not work. In theory you should have a del "event" when hitting shift+backspace. This is not the case in all progs I tested so far (terminal and default text editor with gui).

cat >/etc/ld.so.conf<<END
/lib
/usr/lib
END

ldconfig

When a X window which is displaying GLES2 render output is:
- Either dragged out of the visibile screen region (?)
- Is made larger than it was at the creation time of the GLES2 context
then:
- The performance will drop significantly as the EGL buffer swaps won't occur as fast as before,
- The render output will be corrupted and flicker
- In rare occasions crashes the application or locks up the entire system (which forces you to remove the battery)

The problem can not be solved without restarting the application (Possibly forcing players to complete the same goals again as they were unable to reach a savepoint or similar things).

This was confirmed in Mupen64Plus and the PowerVR GLES Samples and some of my own projects (which are, however, based on the PowerVR samples).
As I didn't see any other reports about this problems (but had it confirmed by other Pandora users too) I would expect that it can be fixed by using a newer X, kernel or combination of them as other OMAP users would be effected too.

Found I'd kept some notes about making the Pandora's wake on alarm function work...

drivers/rtc/rtc-twl4030.c

twl4030_rtc_remove (for when the driver is removed) and twl4030_rtc_
shutdown (for when the system is shutdown) functions need to change. Or at least the shut down does.

oh YUK who's been using goto....

changes in drivers/rtc/rtc-twl4030.c

twl4030_rtc_remove
commented out //mask_rtc_irq_bit(BIT_RTC_INTERRUPTS_REG_IT_ALARM_M);

twl4030_rtc_shutdown
replaced
// mask_rtc_irq_bit(BIT_RTC_INTERRUPTS_REG_IT_TIMER_M |
// BIT_RTC_INTERRUPTS_REG_IT_ALARM_M);
with
mask_rtc_irq_bit(BIT_RTC_INTERRUPTS_REG_IT_TIMER_M);

twl4030_rtc_suspend
replaced
// mask_rtc_irq_bit(BIT_RTC_INTERRUPTS_REG_IT_TIMER_M |
// BIT_RTC_INTERRUPTS_REG_IT_ALARM_M);
with
mask_rtc_irq_bit(BIT_RTC_INTERRUPTS_REG_IT_TIMER_M);

twl4030_rtc_init
added (before return!)
twl4030_rtc_alarm_irq_set_state(true);

Should probably check reg 0x2B with mask 0x08 to see if alarm should be left enabled or not ?
but only in twl4030_rtc_shutdown and ?remove? NOT in twl4030_rtc_init
should remove ever happen ???

I'm guessing it's best to not enable the alarm irq's all the time???

I found a bug when I tried to debug my code with strace.

It happens when strace has to handle conditional SWI's.

Example: (ASM code) save as "condswi.s"
-----
.data
.text
_globl _start
_start:
mov r0,#0 @ return code
mov r7,#1 @ sys_exit
swine 0
swieq 0
.end
-----

The code assembles/links fine using 'as' and 'ld'.
> as condswi.s
> ld a.out -o condswi

When I want to strace the executable file I got the following error.
> strace ./condswi
'syscall: unknown syscall trap 0x1f000000'
As a result strace quits.

Thanks,
Dennis

Since I don't have a Pandora, I used the videos that EvilDragon has shown to discover this issue. Nowhere did I see in the First Boot Wizard that it tells you what the default controls in XFCE does like for example how the mouse moves and clicking using the nubs. It also doesn't tell you what the gaming buttons do (L and R included).

Would be nice to know for first time users. It should also tell a user if they want to change these setting to look at a specific place after the wizard is done. We shouldn't have the users figure the controls and such out by trial and error. When is that ever fun? :)

Trashing a file/folder seems to MOVE it to a certain trash directory on the MAIN VOLUME, rather then the ORIGINATING VOLUME, hence this can result in a COPY RATHER THAN A MOVE operation, if the trashed file originates from a filesystem other than the main filesystem!

This is very inefficient and faulty as it is:
a) Very time consuming
b) And in case of large files this operation can even fail due to not enough free space on the main volume.

populating volatile requires execution of initialization script, which I had to minimally modify for it to start working. I am placing that file to /usr/pandora/init/ folder.

Any suggestions for better place for custom init scripts? :)

This script fixes utmp (and pndnotifyd) problem for systemd image

Noticed problems (TODO):
boot sequence recreates cache, run and tmp dirs before populate-volatile inits them. One of the solutions would be to deploy system with symlinks already in place (pointing to /var/volatile subdirs)

-------------
Deployment:

delete:
/etc/init.d/populate-volatile.sh
/etc/rcS.d/S37populate-volatile.sh
/var/{cache,lock,log,run,tmp}

add:
/etc/systemd/system/populate-volatile.service
/usr/pandora/init/populate-volatile.sh

symlink:
/etc/systemd/system/basic.target.wants/populate-volatile.service to /etc/systemd/system/populate-volatile.service

URLs for files:
http://dl.dropbox.com/u/18781412/next/populate-volatile.service
http://dl.dropbox.com/u/18781412/next/populate-volatile.sh

.desktop files only include the English stuff right now (using the .desktop default tags.)

Be nice if in addition to 'title', we also include 'title[de]' type extras, whatever is in the PXML.xml

pnd_pxml_t includes all the goods, but pnd_disco_t has only the summarized english stuff.

The 'Nubs' application needs a setting for deadzone. Using the nubs in First person shooters the nubs trigger in directions when not even touched. An adjustment for deadzone would eliminate accidental movement caused by vibration.

Also, application profiles for the nubs would be very useful. Different apps use the nubs in different ways and it would be great to not have to go and change settings before launching each application.

Hi there,

Many games depend on sdl-perl. Perl and sdl are part of the base OS, but not not the connection between the two.

Cheers,
sebt3

When I have received my Pandora, I got an SD Card but it was not formated. I didn't have any card reader so I didn't now what was wrong (the Pandora or the SD Card).

When an unformated SD Card is inserted in the Pandora, it should be nice to ask if we want to format it.

That still need to be configurable, but chris_c does a good job I'ld like to see integraded in the main image :
http://www.gp32x.com/board/index.php?/topic/58794-a-script-to-shutdown-after-charging/
and :
http://chrisc.bedroomcoders.co.uk/?p=77

http://www.wader-project.org
http://trac.wader-project.org
http://trac.wader-project.org/wiki/Developers

this would be really usefull to monitor reception/network of mobile broadband dongle and it also counts ur mobile data usage (for those that have limits).

It'll have to be in the SKEL.
/etc/skel/Applications/Settings/Thunar/uca.xml

People can replace the contents of their current uca.xml file with the following to gain this functionality now.
/home/[USERNAME]/Applications/Settings/Thunar/uca.xml

<?xml encoding="UTF-8" version="1.0"?>
<actions>
<action>
<icon>Terminal</icon>
<name>Open Terminal Here</name>
<command>exo-open --working-directory %f --launch TerminalEmulator</command>
<description>Example for a custom action</description>
<patterns>*</patterns>
<directories/>
</action>
<action>
<icon>emblem-symbolic-link</icon>
<name>Mount PND</name>
<command>/usr/pandora/scripts/pnd_run.sh -p %f -m &amp;&amp; thunar /mnt/utmp/</command>
<description></description>
<patterns>*.pnd;*.PND</patterns>
<other-files/>
<text-files/>
</action>
<action>
<icon>gtk-jump-to-rtl</icon>
<name>Unmount PND</name>
<command>/usr/pandora/scripts/pnd_run.sh -p %f -u</command>
<description></description>
<patterns>*.pnd;*.PND</patterns>
<other-files/>
<text-files/>
</action>
</actions>

PROBLEM DESCRIPTION: In many GUI dialogues the default action button (OK, ACCEPT, YES, …) cannot be triggered with a simple key press. Pressing ENTER triggers the element which has focus (usually the first dialogue element), but not the default action, as most computer users would expect!

MY WORKAROUND MEANWHILE: Press ALT plus the underlined letter of your desired action. But that's not very convenient as the underlined letter needs to be looked upon first, as the button label can differ from situation to situation (ACCEPT, OK, YES, …) and then 2 keys need to be pressed.

SUGGESTED SOLUTION: The custom of many operating systems should be adapted on the OpenPandora as well. SPACE triggers the currently focused element (whatever that may be: list item, radio button, checkbox, etc), and ENTER triggers the default action of that dialogue.
Or even better use the Pandora specific A/B/Y/X keys cleverly in those dialogues/selections. I don't know whether there is a standard functionality assignment within this key-group, in MiniMenu "B" is start/confirm, I do not know of any other standard assignments yet.

I am suggesting the new option "Grid stop horizontal" for MiniMenu with the following possible settings and their effect.

Yes -> If the boundary is reached the focus remains there.
No -> If the boundary is reached the focus starts at the opposite side again.
Jump to next/prev line -> If the boundary is reached the focus jumps to the next/previous line and the opposite side.

If you set MiniMenu's option "Live (not exit) on app run?" to YES, and you quit an application and return to the still running MiniMenu, it runs in windowed mode, and takes about 2 seconds until it goes into full screen mode again.

This delay should at best be not noticeable at all, at most 0-1 seconds.

ISSUE ENVIRONMENT:

If you hold your Pandora in your hands, you only have your thumbs for typing, hence your typing abilities are quite limited.

It is hard to press more than 2 keys, and hard to press 2 keys on the same side, as your second thumb does not reach the opposite side all too easy (except you have a basketball player's hands).

The modifier keys Start/ALT Select/CTRL Pandora/(META, I guess?) are in the middle of the keyboard, hence the can easily pressed together with another key. But SHIFT lies at an ergonomically problematic side (left boundary). Hence hard to press with other left side keys.

SUGGESTED SOLUTION:

I suggest that Pandora OS offers an optional and configurable input help.

[1|2|3 presses | hold for duration x] SHIFT [timeout y] [modifier key 2] [timeout z] key

Produces the same result like SHIFT + modkey2 + key

Very likely modifier key 2 and timeout z are not necessary, as you can trigger the SHIFT hold (by your defined action), and then press the modifier key 2 and the other key at once, as the other modifier keys are laid at a central (ergonomically better) position.

Imagine this tab content:
Nubs, Pandora Input, PNDstore.

CURRENT BEHAVIOR:

Pressing "P" and "N" is succession jumps to "Nubs" eventually, as it first jumps to the first "P" item ("Pandora Input"), then to the first "N" item, "Nubs".

DESIRED BEHAVIOR:

That it acts like in most Linux file browsers or the Mac OS X Finder.

In detail: Pressing "P" and "N" within a certain time limit should combine them to the string "PN" and jump to the first item which starts with that string, i.e. "PNDStore". If no item matches, reduce the combination string by one character, try to match again, if that doesn't match, try matching again with one character less, ... , until only one character is left, and if that one doesn't match jump to the nearest previous lexicographical character, which matches.

The command line environment of the default OS does not include the command "man" and the respective manpages of the installed programs.
I personally often use manpages, as I hardly remember the syntax/arguments of all programs I use.
Having them would be very convenient!

EVALUATING NEEDED STORAGE AMOUNT

Some may argue that storage space is quite limited on the NAND.

Yes I agree. Hence the number of included programs is limited too. Hence it would not be all too many manpages.

When I press TAB into an empty Terminal prompt, I get 1406 possibilities, which is about the amount of available commands/programs if we ignore aliases, etc.

Multiply that with an average of 20 kb per manpage, and you get about 30 MB in total. With compression this could possibly be brought down to 15-20 MB.

POSSIBLE SOLUTIONS

A) Store into the NAND the manpages of core OS CLI programs only, and not all the library/system/etc documentation.

B) Include only the man command into the core OS, and compile/configure it in such a fashion, that it finds the manpages within non-NAND media (SD card, USB volume, etc).
b1) Specify that it simply looks into a certain path within the /media/*/pandora/ structure.
b2) Or offer PND packages like man-core.pnd, man-extended.pnd,... you get the idea or even man-custom.pnd (which would look into its appdata or a certain path for custom added manpages, to somehow get the b1 approach within the PND approach).

a) Host multiple volumes at once.
b) Select one for hosting first, later host additional ones. (I guess this scenario is more complicated)